[gmx-developers] GROMACS performance on Opteron Dual Core processors!

David van der Spoel spoel at xray.bmc.uu.se
Mon Oct 2 20:19:44 CEST 2006

Kazem Jahanbakhsh wrote:
> Dear Christian,
> I have confused with your description. I have some simple 
> questions now, first is what does "Opteron-DC" stand for? 
> Doesn't it mean "Dual-Core Opteron"? You stated that "N" 
> column in the benchmark table means number of cores and not 
> the processors. I don't realize this at all. In the 
> presented benchmark table for single processors they brought
> the results of "Opteron" processors with N=2 and N=4, if we
> consider the "N" value as the number of cores on one 
> hardware chip, this does not match with the the Opteron 
> processors that exist in single core architecture in the 
> real world. I can't understand this contradiction in any 
> way. Also in that table they showed the performance outputs
> of "Opteron-DC" processors with N=2 and N=4. This results 
> doesn't mean if we consider the "N" as the number of cores 
> per chip because in the market the "Dual-Core Opteron" 
> processors only exist in dual core
>  structure and AMD doesn't
> bring the Quad-Core processors into the market. for this 
> reason I cant interpret the "N" number as the number of 
> cores on single processor. Please clarify me with this 
> misunderstanding.
> Sorry for the confusion. N is the number of cores, so N is 1 for an old 
opteron, if N is 4 it means 2 dual-core chips.

One new core is faster than one old opteron.

David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
Dept. of Cell and Molecular Biology, Uppsala University.
Husargatan 3, Box 596,  	75124 Uppsala, Sweden
phone:	46 18 471 4205		fax: 46 18 511 755
spoel at xray.bmc.uu.se	spoel at gromacs.org   http://folding.bmc.uu.se

More information about the gromacs.org_gmx-developers mailing list